← All comparisons
Hermes 4 - Llama-3.1 70B (Reasoning) vs Claude 4.1 Opus (Reasoning)
Nous Research vs Anthropic — side-by-side benchmark comparison
| Hermes 4 - Llama-3.1 70B (Reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 16.0 | 42.0 |
| Coding Index | 14.4 | 36.5 |
| Math Index | 68.7 | 80.3 |
| Output speed (tok/s) | 92.8 | 44.5 |
| Blended price ($/1M) | $0.20 | $32.81 |
| Time to first token (s) | 0.64s | 8.55s |
| aime | — | — |
| aime 25 | 68.7% | 80.3% |
| artificial analysis coding index | 14.40 | 36.50 |
| artificial analysis intelligence index | 16.00 | 42.00 |
| artificial analysis math index | 68.70 | 80.30 |
| gpqa | 69.9% | 80.9% |
| hle | 7.9% | 11.9% |
| ifbench | 31.3% | 55.4% |
| lcr | 6.7% | 66.3% |
| livecodebench | 65.3% | 65.4% |
| math 500 | — | — |
| mmlu pro | 81.1% | 88.0% |
| scicode | 34.1% | 40.9% |
| tau2 | 22.5% | 71.4% |
| terminalbench hard | 4.5% | 34.3% |
Benchmark data from Artificial Analysis.